Upload
giancarlotorr1548
View
244
Download
0
Embed Size (px)
Citation preview
8/3/2019 Shelf Life of Package Foods
1/25
CHAPTER 13
Shelf Life of Packaged Foods, ItsMeasurement and Prediction
GORDON L. ROBERTSON
Because all foods change during distribution, usually deteriorating, an understanding
of and ability to predict these processes are indispensable components of food product
development. It is the food product developer/processors responsibility to comprehend
the changes that take place in distribution. Shelf life prediction consists of two-parts:quantifying the inherent product deterioration characteristics and coupling those with
the properties of the packaging/distribution system. Mathematical models have been
developed and used with some effectiveness as a guide to shelf life. To reduce the
probability of surprises, however, actual shelf life testing is essential. Whether acceler-
ated shelf life testing may be applicable depends on the product and its packaging.
THE quality of most foods and beverages decreases with time. Exceptions
include distilled spirits that develop desirable flavors during storage inwooden barrels, some wines that increase in flavor complexity in bottles, and
many cheese varieties where aging leads to desirable flavors and textures.
For the majority of foods and beverages, however, a finite time occurs
before the product becomes unacceptable. This time from production to unac-
ceptability is usually designated shelf life.
No simple, generally accepted definition of shelf life exists. The Institute
of Food Technologists (Anonymous, 1974) has defined shelf life as the period
between the manufacture and the retail purchase of a food product, duringwhich time the product is of satisfactory quality in terms of nutritional value,
flavor, texture, and appearance. An alternative definition is that shelf life is
This material is excerpted from Robertson, Gordon L., Food Packaging: Principles and
Practices, Copyright 1993, Marcel Dekker, Inc., New York. Used with permission.
2000 by CRC Press, Inc.
8/3/2019 Shelf Life of Package Foods
2/25
that period between the packing of a product and its use, for which the quality
of the product remains acceptable to the product user.
Shelf life must be determined for each product by the processor. Storage
studies are an indispensable element of food product development, with the
processor attempting to provide the longest shelf life practicable consistentwith economics and distribution. Inadequate shelf life will lead to consumer
dissatisfaction and complaints, and eventually adversely affect the acceptance
and sales of branded products.
Since the advent of the consumer movement, many different types of open
dating systems have been proposed as part of the consumers right to know.
An open date on a food product is a legible, easily read date which is displayed
on the package with the purpose of informing the consumer about the shelf
life of the product. Several types of dates can be used (Dethmers, 1979;
Labuza, 1982):
T packdate: the date on which the product was packed into its primary
package (it does not provide any specific information as to the quality
of the product when purchased or how long it might retain its quality
after purchase.)
T display date: the date on which the product was placed on the shelf by
the retailer
T
pull or sell by date: the last date on which the product should be soldin order to allow the consumer a reasonable length of time in which to
use it
T best before or best if used by date: the last date of maximum high
quality
T use by date or expiration date: the date after the which the food
should no longer be at an acceptable level of quality
These forms of dating are used infrequently because quality changes gener-
ally occur slowly and it is really not possible to state that a food will beacceptable one day and unacceptable the next.
FACTORS CONTROLLING SHELF LIFE
Product shelf life is controlled by three factors:
T product characteristicsT the environment to which the packaged product is exposed during
distribution
T the properties of the package
Product shelf life may be altered by changing its composition and form,
2000 by CRC Press, Inc.
8/3/2019 Shelf Life of Package Foods
3/25
the environment to which it is exposed, or its packaging system (Harte and
Gray, 1987).
PRODUCT CHARACTERISTICS
Perishability
Based on the nature of the changes that can occur during storage, foods
may be divided into three categories: perishable, semiperishable, or ambient
temperature shelf stable, which translate into very short shelf life products,
short to medium shelf life products, and medium to long shelf life products.
Perishable foods are those subject to microbiological and/or enzymatic
deterioration and so must be held at chill or freezer temperatures. Examplesof such foods would include milk; fresh meat, poultry, and fish; and many
fresh fruits and vegetables.
Semiperishable foods are those that contain natural inhibitors (e.g., some
cheeses, eggs, etc.) or those that have received minimal preservation treatment
(e.g., milk pasteurization, ham smoking, and vegetable fermentation) that
delivers greater tolerance to environmental conditions and abuse during distri-
bution.
Ambient temperature shelf stable foods are often regarded as nonperish-able at room temperatures. Some natural foods fall into this category
(e.g., cereal grains and nuts, and some confectionery products). Processed
food products can be shelf stable if they are preserved by thermal sterilization
(e.g., canned foods), contain preservatives (e.g., soft drinks), are formulated
as dry mixes (e.g., cake mixes), or processed to reduce their water content
(e.g., raisins or crackers). However, ambient temperature shelf stable foods
only retain this status if the integrity of the package which contains them
remains intact. Even then, their shelf life is finite due to deteriorative chemical
reactions that proceed at ambient temperature and the permeation through
packages of gases, odors, and water vapor.
Bulk Density
For packages of similar shape, equal weights of products of different bulk
densities have different free space volumes and, as a consequence, package
areas and package behavior differ. This has important implications whenchanges are made in package size for the same product, or process alterations
are made, resulting in changes to the product bulk density.
The bulk density of food powders can be affected by processing and packag-
ing. Some food powders (e.g., milk and coffee) are instantized by treating
individual particles so that they form free-flowing agglomerates or aggregates
2000 by CRC Press, Inc.
8/3/2019 Shelf Life of Package Foods
4/25
in which there are relatively few points of contact; the surface of each particle
is thus more easily wetted when the powder is rehydrated.
The free space volume has an important influence on the rate of oxidation
of foods, because if a food is packaged in air, a large free space volume is
undesirable since it constitutes a large oxygen reservoir. Conversely, if theproduct is packaged in an inert gas, a large free space volume acts as a large
sink to minimize the effects of oxygen transferred through the film. It
follows that a large package area and a low bulk density result in greater
oxygen transmission.
Concentration Effects
The progress of a deteriorative reaction in a packaged food can be monitored
by following the changes in concentrations of some key components. In many
foods, however, the concentration varies from point to point, even at zero
time. Because most of these compounds have little opportunity to move, the
concentration differences increase as the reactions proceed out from isolated
initial foci.
Further, several different deteriorative reactions may proceed simultane-
ously, and different stages may have different dependence on concentration
and temperature. Such a situation is frequently the case for chain reactions
and microbiological growth which have both a lag and a log phase with verydifferent rate constants.
Thus, for many foods it may be difficult to obtain kinetic data useful for
predictive purposes. Sensory panels to determine the acceptability of the food
are therefore the recommended procedure.
DISTRIBUTION ENVIRONMENT
Climatic
The deterioration in product quality of packaged foods is often closely
related to the transfer of mass and heat through the package. Packaged foods
may lose or gain moisture; they will also reflect the temperature of their
environment because very few food packages are good thermal insulators.
Thus, the distribution environment has an important influence on the rate of
deterioration of packaged foods.
Mass Transfer
With mass transfer, the exchange of vapors and gases with the surrounding
atmosphere is of primary concern. Water vapor and oxygen are generally of
most importance, although the exchange of volatile aromas from or to the
2000 by CRC Press, Inc.
8/3/2019 Shelf Life of Package Foods
5/25
product from the surroundings can be important. Transmission of nitrogen
and carbon dioxide may have to be taken into account in some packages.
Generally, the difference in partial pressure of the vapor or gas across the
package barrier controls the rate and extent of permeation, although transfer
can also occur due to the presence of pinholes in the material, channels inseals and closures, or cracks that result from flexing of the package material
during filling and subsequent handling. Because the gaseous composition of
the atmosphere is constant at sea level, the partial pressure difference of gases
across the package material depends on the internal atmosphere of the package
at the time the package was sealed.
In contrast to the common gases, the partial pressure of water vapor in the
atmosphere varies continuously, although the variation is generally much less
in controlled climate environments (Porter, 1981). Thus, mass transfer depends
on the partial pressure difference across the package barrier, and on the nature
of the barrier itself.
Heat Transfer
One of the major determinants of product shelf life is the temperature to
which the product is exposed during its lifetime. Without exception, food prod-
ucts are exposed to fluctuating temperature environments, and it is important,
if an accurate prediction of shelf life is to be made, that the nature and extent ofthese temperature variations are known. There is little point in carefully control-
ling the processing conditions inside the factory and then releasing the product
into the distribution and retail system without some knowledge of the conditions
it will experience in that system. The storage climates inside warehouses and
supermarkets are only broadly related to the external climate.
If the major deteriorative reactions causing end of shelf life are known,
expressions can be derived to predict the extent of deterioration as a function
of available time-temperature storage conditions.Fundamental to a predictive analysis is that the particular food under consid-
eration follows the laws of additivity and commutativity. Additivity
implies that the total extent of the degradation reaction in the food produced
by a succession of exposures at various temperatures is the simple sum of the
separate amounts of degradation, regardless of the number or spacing of each
time-temperature combination. Commutativity means that the total extent of
the degradation reaction in the food is independent of the order of presentation
of the various time-temperature experiences.
Shelf Life Plots
One useful approach to quantifying the effect of temperature on food quality
is to construct shelf life plots (Labuza and Kamman, 1983). Several models
2000 by CRC Press, Inc.
8/3/2019 Shelf Life of Package Foods
6/25
are in use to represent the relationship between the rate of a reaction (or the
reciprocal of rate which can be time for a specified loss in quality or shelf
life) and temperature. The two most-used models are the Arrhenius and linear,
and these are shown in Figure 13.1.
The equations for these two plots are
s = o expEA
R1Ts
1
To (13.1)
and
s = o eb(TsTo) (13.2)
where s = shelf life at temperature Ts and o = shelf life at temperature To.
If only a small temperature range is used (less than 40F), there is little
error in using the linear plot rather than the Arrhenius plot.
Most deteriorative reactions in foods can be classified as either zero or first
order. The way in which these two reaction orders can be used to predict the
extent of deterioration as a function of temperature is outlined.
Zero-Order Reaction Prediction
The change in a quality factor A when all extrinsic factors are held constant
is expressed in Equation (13.3):
Ae = Ao kzs (13.3)
FIGURE 13.1 (a) Arrhenius plot of log shelf life (s) versus reciprocal of the absolute temperature
(K) showing a slope ofEA/R, and (b) linear plot of log shelf life versus temperature (C) showing
a slope ofb. Reprinted from Robertson (1993), p. 345, by courtesy of Marcel Dekker, Inc.
2000 by CRC Press, Inc.
8/3/2019 Shelf Life of Package Foods
7/25
and
Ao = Ae = kzs (13.4)
where Ae = value ofA at end of shelf life, Ao = value ofA initially, kz = zeroorder rate constant (time1), and s = shelf life in days, months, years, etc.
For variable time-temperature storage conditions, Equation (13.3) can be
modified as follows:
Ae = Ao (Kii) (13.5)
where kii = the sum of the product of the rate constant ki at each temperature
and Ti times the time interval i at the average temperature Ti for the giventime period .
To apply this method, the time-temperature history is divided into suitable
time periods and the average temperature in each time period is determined.
The rate constant for each period is then calculated from the shelf life plot
using a zero-order reaction. The rate constant is multiplied by the time interval
i, and the sum of the increments ofkii gives the total amount lost at any time.
Alternatively, instead of calculating actual rate constants, the time for the
product to become unacceptable (i.e., for A to become Ae) can be measured,
and Equation (13.5) modified to give
fc = fraction of shelf life consumed
= change in A divided by total possible change in A
=Ao A
Ao Ae(13.6)
=(kii)
(kis)(13.7)
= isTi (13.8)
The temperature history is divided into suitable time periods and the averagetemperature Ti at each time period evaluated. The time held at that temperature
i is then divided by the shelf life s for that particular temperature, and the
fractional values summed to give the fraction of shelf life consumed.
The shelf life can also be expressed in terms of the fraction of shelf life
remaining, fr:
2000 by CRC Press, Inc.
8/3/2019 Shelf Life of Package Foods
8/25
fr = 1 fc (13.9)
Thus for any temperature Ts:
frs = (1 fc)s = shelf life at temperature Ts (13.10)
In other words, the shelf life at any temperature is the fraction of shelf life
remaining times the shelf life at that temperature.
The above method is referred to as the TTT or time/temperature tolerance
approach (Van Arsdel, 1969). To use this method, the period of time (desig-
nated as the high quality life or HQL) for 70 to 80% of a trained sensory
panel to correctly identify the control samples from samples stored at various
other temperatures using the triangle or duo-trio difference test is determined.
The change in quality at this stage has been designated the just noticeable
difference (JND). The HQL has no real commercial significance and is quite
different from the practical storage life (PSL), which is of interest to food
processors and consumers. The ratio between PSL and HQL is often referred
to as the acceptability factor and can range from 2:1 up to 6:1.
Generally, the HQL varies exponentially with temperature. When overall
quality rather than just one single quality factor is measured, however, a semi-
logarithmic plot results in curved rather than straight lines.Time/temperature tolerance relationships are not strict mathematical func-
tions but empirical data subject to large variability, particularly because of
variations in product, processing methods, and packaging (the PPP factors).
Therefore, any shelf life prediction made will be specific for a particular
product processed, packaged, and stored under specific conditions. Predictions
cannot be made with any precision on the quality or quality change in a food
from knowledge of its time-temperature history and TTT literature data only.
Therefore, in determining the shelf life of foods, the PPP factors in additionto the TTT relationships must be taken into account.
Rather than follow the TTT approach and use a linear model to relate the
quality loss with temperature, a distribution system using the Arrhenius model
and converting a variable temperature history to equivalent time at a standard
temperature may be used (Rosenfeld, 1984).
The distribution system is divided into four stages and tables developed
describing the distribution system as equivalent time for a range ofQ10 values.
Data are also collected that enable calculation of the mean and standarddeviation days that a product spends at each stage of the distribution system.
If the failure point of the product is beyond the 90 or 95 percentile, then the
product is considered to have sufficient shelf life to survive the given distribu-
tion system. This information could also be used as a guide for stability limits
for new product development.
2000 by CRC Press, Inc.
8/3/2019 Shelf Life of Package Foods
9/25
First-Order Reaction Prediction
The expression for a first-order reaction for the case in which all extrinsic
factors are held constant is shown in Equation (13.11).
Ac = Ao exp(ks) (13.11)
From this an expression can be developed to predict the amount of shelf
life used up as a function of variable temperature storage for a first-order
reaction in the form:
A = Ao exp(kii) (13.12)
where A = the amount of some quality factor remaining at the end of the time-
temperature distribution, and kii has the same meaning as in Equation (13.5).
If the shelf life is based simply on some time to reach unacceptability,
Equation (13.12) can be modified to give an analogous expression to that
derived for the TTT method. Note that because of the exponential loss of
quality, Ae will never be zero. Thus,
1nA
Ao= kii (13.13)
and
ki =1n Ae/Ao
s(13.14)
where ln A/Ao = fraction of shelf life consumed at time and ln Ae/Ao =
fraction of shelf life consumed at time s.
The fraction of shelf life remaining, fr, is
fr = 1 1n Ao/A
1n Ao/Ae= 1 i
sTi (13.15)
Sequential Fluctuating Temperatures
Although the above analysis can be applied to any random time/temperature
storage regime, in practice many products are exposed to a sequential regular
fluctuating temperature profile, especially if held in trucks, rail cars and
2000 by CRC Press, Inc.
8/3/2019 Shelf Life of Package Foods
10/25
warehouses. This is because of the daily day-night pattern resulting from
exposure to solar radiation. Many of these patterns can be assumed to follow
either a square or sine wave form as shown in Figure 13.2.
Equations have been developed (Labuza, 1979) for both zero and first-order
reactions that enable calculation of the extent of a degradative reaction for afood subjected to either square wave or sine wave temperature functions. The
extent of reaction after a time period is the same as it would have been if the
food had been held at a certain steady effective temperature for the same
length of time. This effective temperature is higher than the arithmetic mean
temperature. Comparisons for losses in a theoretical temperature distribution
show that for less than 50% degradation the losses are about the same for
zero and first order at any time, and thus determination of the reaction order
is not critical. However, the temperature sensitivity (Q10) of the reaction is
very important in making predictions.
Simultaneous Mass and Heat Transfer
In the majority of distribution environments, many packaged foods undergo
changes in both moisture content and temperature during storage as a result
of varying temperature and relative humidity conditions in the environment.
This has the effect of complicating the calculations for prediction of the
shelf life of packaged foods. It is unlikely that a package would be totally
FIGURE 13.2 Square and sine wave temperature fluctuations of packaged foods where ao is
the amplitude. Reprinted from Robertson (1993), p. 352, by courtesy of Marcel Dekker, Inc.
2000 by CRC Press, Inc.
8/3/2019 Shelf Life of Package Foods
11/25
impermeable to water vapor, and therefore the aw would change with time.
This complicates the calculation of quality loss, since the rate is now dependent
on both temperature and aw.
A further complication is that data on the relative humidity distribution of
environments in which foods are stored are scarce and not as easily predictedas the external temperature distribution. Therefore, prediction of the actual
shelf life loss of packaged foods will only be approximate. More complete
data are required about the humidity distribution of food storage environments
so that shelf life predictions can be further refined.
PACKAGE PROPERTIES
Foods can be classified according to the amount of protection required,
as shown in Table 13.1. The advantage of this sort of analysis is that
attention can be focused on the key requirements of the package such as
maximum moisture gain or oxygen uptake. This then enables calculations
to be made to determine whether or not a particular package structure
would provide the necessary barrier required to give the desired product
shelf life. Metal cans and glass bottles or jars can be regarded as essentially
impermeable, while paper-based packaging materials can be regarded as
permeable. Plastic-based packaging materials provide varying degrees of
protection, depending largely on the nature of the polymers and theirpackage structures.
The expression for the steady state permeation of a gas or vapor through
a thermoplastic material can be written as (Robertson, 1993):
w
t=
P
XW A W (p1 p2) (13.16)
where P/X is the permeance (the permeability constant P divided by thethickness of the film X), A is the surface area of the package, p1 and p2 are
the partial pressures of water vapor outside and inside the package, and w/
tis the rate of gas or vapor transport across the film, the latter term correspond-
ing to Q/tin the integrated form of the expression.
Water Vapor Transfer
The prediction of the moisture transfer either to or from a packaged foodbasically requires analysis of Equation (13.16) given certain boundary condi-
tions. The simplest analysis requires the assumptions that P/Xis constant, that
the external environment is at constant temperature and humidity, and that
p2, the vapor pressure of the water in the food, follows some simple function
of the moisture content.
2000 by CRC Press, Inc.
8/3/2019 Shelf Life of Package Foods
12/25
TABLE 13.1. Degree of Protection Required by Various Foods andBeverages [Assuming One Year Shelf Life at 25C (79F)].
Requires
Maximum Good
Amount of Other Gas Maximum Requires Barrier to
Food/ O2 Gain Protection Water Gain High Oil Volatile
Beverage (ppm) Needed or Loss Resistance Organics
Canned milk 15 No 3% Loss Yes No
and meats
Baby foods 15 No 3% Loss Yes Yes
Beers and 15
8/3/2019 Shelf Life of Package Foods
13/25
The critical point about Equation (13.16) is that the internal water vapor
pressure is not constant but varies with the moisture content of the food at
any time. Thus the rate of gain or loss of moisture is not constant but falls
as p gets smaller. Therefore some function ofp2, the internal vapor pressure,
as a function of the moisture content, must be inserted into the equation tobe able to make proper predictions. If a constant rate is assumed, the product
will be overprotected.
In low and intermediate moisture foods, the internal vapor pressure is
determined solely by the water sorption isotherm of the food. Several functions
can be applied to describe a sorption isotherm, although the preferred one is
the G.A.B. (from Guggenheim-Anderson-de Boer) model (Van Den Berg and
Bruin, 1981). If a linear model is used, the result is directly integratable, but if
the G.A.B. model is used, it must be numerically evaluated using computational
techniques.
In the simplest case, the isotherm is treated as a linear function:
m = b aw + c (13.17)
where m = moisture content in g H2O per g solids, aw = water activity, b =
slope of curve, and c = constant.
The moisture content can be substituted for water gain using the relationship:
m =W(weight of water transported
Ws (weight of dry solids enclosed)(13.18)
W= mWs (13.19)
and
W= mWs (13.20)
By substitution:
W
t=mWs
t=
P
XW A W pome
b
pom
b (13.21)
which on rearranging gives
m
me m=
P
XW
A
WsW
po
bW t (13.22)
2000 by CRC Press, Inc.
8/3/2019 Shelf Life of Package Foods
14/25
and on integrating
lnme mi
me m=
P
XW
A
WsW
po
bW t (13.23)
where me = equilibrium moisture content of the food if exposed to external
package RH, mi = initial moisture content of the food, m = moisture content
of the food at time t, and po = water vapor pressure of pure water at the storage
temperature (notthe actual vapor pressure outside the package).
A plot of the log of the unaccomplished moisture change (the term on the
left-hand side of Equation (13.23)) versus time is a straight line with a slope
equivalent to the bracketed term on the right-hand side of the equation.
The end of product shelf life is reached when m = mc, the critical moisturecontent, at which time t = s, the shelf life. Thus Equation (13.23) can be
rewritten as
lnme mi
me mc=
P
XW
A
WsW
po
bW s (13.24)
The relationship between the initial, critical, and equilibrium moisture con-
tents is illustrated in Figure 13.3.
To simplify matters, the packaging parameters can be combined into one
constant as
FIGURE 13.3 Typical moisture sorption isotherm for a food product in which mi = initial
moisture content; mc = critical moisture content of product; me = equilibrium moisture content.
Reprinted from Robertson (1993), p. 358, by courtesy of Marcel Dekker, Inc.
2000 by CRC Press, Inc.
8/3/2019 Shelf Life of Package Foods
15/25
=P
XW
A
Ws(13.25)
Using Equation (13.25), one can calculate a minimum , given a critical
moisture content and maximum desired shelf life. Then from Equation (13.24)for a given package size and weight of product, the permeance can be calculated
and a package structure to satisfy this condition selected.
Equation (13.23) and the corresponding one for moisture loss:
lnmi me
m me=
P
XW
A
WsW
po
bW t (13.26)
have been extensively tested for foods and found to give excellent predictionsof actual weight gain or loss (Labuza, 1984). These equations are also useful
when calculating the effect of changes in the external conditions (e.g., tempera-
ture and humidity), the surface area-to-volume ratio of the package, and
variations in the initial moisture content of the product.
Given specific external conditions and a critical aw for moisture gain, the
shelf life is
s = W WsA
= W VA
= W r (13.27)
where , , and are constants proportional to
ln me mime m
PX
W
po
b (13.28)
where Ws = weight of food solids = p V, p = density of food, V= volumeof food, r= characteristic package thickness, and s = time to end of shelf life.
Because the V/A ratio decreases as package size gets smaller by a factor
equivalent to the characteristic thickness of the package, the shelf life using
the same film will decrease directly by this thickness. Thus, to ensure adequate
shelf life for a food in varying sizes of packages, shelf life tests should be
based on the smallest package.
Gas and Odor Transfer
The gas of major importance in packaged foods is oxygen since it plays a
crucial role in many reactions which affect the shelf life of foods, e.g., microbial
growth, color changes, oxidation of lipids and consequent rancidity, and senes-
cence of fruit and vegetables.
2000 by CRC Press, Inc.
8/3/2019 Shelf Life of Package Foods
16/25
The transfer of gases and odors through packaging materials can be analyzed
in an analogous manner to that described for water vapor transfer, provided
that values are known for the permeance of the packaging material to the
appropriate gas, and the partial pressure of the gas inside and outside the
package.Packaging can control two variables with respect to oxygen, and these can
have different effects on the rates of oxidation reactions in foods (Karel,
1974):
T Total amount of oxygen present. This influences the extent of the
reaction and, in impermeable packages, where the total amount of
oxygen available to react with the food is finite, the extent of the
reaction cannot exceed the amount corresponding to the complete
exhaustion of the oxygen present inside the package at the time of
sealing. This may or may not be sufficient to result in an unacceptable
product quality after a period of time, dependent on the rate of the
oxidation reaction. Such a rate is, or course, temperature dependent.
With permeable packages (e.g., plastic packages), where ingress of
oxygen occurs during distribution, two factors are important: sufficient
oxygen may be present inside the package to cause product
unacceptability when it has all reacted with the food, or there may be
sufficient transfer of oxygen into the package over time to result in
product unacceptability through oxidation.
T Concentration of oxygen in the food. In many cases, relationships
between the oxygen pressure in the space surrounding the food and
the rates of oxidation reactions can be established. If the food itself is
very resistant to diffusion of oxygen, then it will probably be very
difficult to establish a relationship between the oxygen pressure in the
space surrounding the food and the concentration of oxygen in the
food.
The principal difference between predominantly water vaporsensitive and
oxygen-sensitive foods is in the fact that the latter are generally more sensitive
by 2 to 4 orders of magnitude (Heiss, 1980). Thus, the amount of oxygen
present in the air-filled headspace of oxygen-sensitive foods must not be
neglected when predicting their shelf life. This amount is actually 32 times
higher per unit volume of air than per unit volume of oxygen-saturated water. A
further complicating factor with oxygen-sensitive foods is that a concentrationgradient occurs in them much more frequently than in moisture-sensitive
foods.
Prediction of the shelf life of food products which deteriorate by two or
more mechanisms simultaneously is more complex. Some general approaches
that can be applied to any number of deteriorative reactions have been proposed
2000 by CRC Press, Inc.
8/3/2019 Shelf Life of Package Foods
17/25
for food such as chips which deteriorate by two mechanisms simultaneously:
oxidation due to ingress of atmospheric oxygen, and loss of crispness due to
ingress of moisture (Quast and Karel, 1972).
The simultaneous transfer of water vapor and gases through the package
when in an environment with fluctuating temperature and humidity makesquantitative analysis of the deteriorative reactions occurring in the foods (and
hence prediction of shelf life) exceedingly complex. Reliance is being placed
on the use of accelerated shelf life testing (ASLT) procedures as a more cost-
effective and simpler method for the determination of product shelf life.
SHELF LIFE TESTING
Generally the shelf life testing of food products falls into one of three
categories (Gacula, 1975):
T experiments designed to determine the shelf life of existing products
T experiments designed to study the effect of specific factors and
combinations of factors such as storage temperature, package
materials, or food additives on product shelf life
T tests designed to determine the shelf life of prototype or newly
developed products
Basic approaches to determining the shelf life of a food product include
T Literature study: the shelf life of an analogous product has been
recorded in the published literature or organization files or reports.
T Turnover time: the average length of time which a product spends in
distribution is found by monitoring sales and, from this, the required
shelf life is estimated. This does not give the true shelf life of theproduct but rather the required shelf life, it being implicitly
assumed that the product is still acceptable for some time after the
average period on the retail shelf.
T End point study: random samples of the product are purchased from
distribution channels and tested in the laboratory to determine their
quality; from this a reasonable estimate of shelf life can be obtained
since the product has been exposed to actual environmental stresses
encountered during distribution.T Accelerated shelf life testing: laboratory studies are undertaken during
which environmental conditions are accelerated by a known factor so
that the product deteriorates at a faster than normal rate. This method
requires that the effect of environmental conditions on product shelf
life can be quantified.
2000 by CRC Press, Inc.
8/3/2019 Shelf Life of Package Foods
18/25
Regardless of the method chosen or the reasons for its choice, sensory
evaluation of the product is likely to be used either alone or in combination
with instrumental analyses to determine the quality of the product. Because
human judgment is the ultimate arbiter of food acceptability, it is essential
that the results obtain from any instrumental or chemical analysis correlateclosely with the sensory judgments for which they are to substitute.
In chemical and physical tests, analytical parameters are isolated so that a
single signal is monitored, whereas sensory responses are more complex
because of the integration of multiple signals due to the interdependence of
appearance, texture, aroma, and flavor of a food. Hedonic responses and, to
a lesser extent, intensity judgments are subject to many experimental influences
such as past exposure to the product and those created by the actual test
protocol.
Difference methods are used to measure whether reference samples are
different from stored samples, or control samples from test samples. These
methods require trained or experienced panelists. Three experimental designs
are commonly used for the purpose of shelf life testing (Labuza and Schmidl,
1988): the paired comparison test, the duo-trio test, and the triangle test.
Further details about these tests can be found inChapter 14 of this book.
In shelf life testing there can be one or more criteria which constitute sample
failure. One criterion is an increase or decrease by a specified amount in the
mean panel score. Another criterion is microbiological deterioration of thefood to an extent that renders it unsuitable or unsafe for human consumption.
Finally, any physical changes such as changes in color, mouthfeel, flavor,
etc., that render the sample unacceptable to either the panel or the consumer
are criteria for product failure. Thus sample failure can be defined as the
condition of the product that exhibits either physical, chemical, microbiologi-
cal, or sensory characteristics that are unacceptable to the consumer, and the
time required for the product to exhibit such conditions is the shelf life of the
product.However, a fundamental requirement in the analysis of data is knowledge
of the statistical distribution of the observations, so that the mean time to failure
and its standard deviation can be accurately estimated, and the probability of
future failures predicted. The length of shelf life for food products is usually
obtained from simple averages of time to failure on the assumption that the
failure distribution is symmetrical (Gacula and Kubala, 1975). If the distribu-
tion is skewed, estimates of the mean time to failure and its standard deviation
will be biased. Further, when the experiment is terminated before all thesamples have failed, the mean time to failure based on simple averages will
be biased because of the inclusion of unfailed data.
In order to improve the method of estimation of shelf life, knowledge of
the statistical distribution of shelf life failures is required, together with an
appropriate model for data analysis.
2000 by CRC Press, Inc.
http://tx67784_14.pdf/http://tx67784_14.pdf/http://tx67784_14.pdf/8/3/2019 Shelf Life of Package Foods
19/25
One problem with shelf life testing is to develop experimental designs
which minimize the number of samples required, thus minimizing the cost of
the testing while still providing reliable and statistically valid answers.
ACCELERATED SHELF LIFE TESTING (ASLT)
BASIC PRINCIPLES
The basic assumption underlying accelerated shelf life testing (ASLT) is
that the principles of chemical kinetics can be applied to quantify the effects
which extrinsic factors such as temperature, humidity, gas atmosphere, and
light have on the rate of deteriorative reactions. By subjecting the food tocontrolled environments in which one or more of the extrinsic factors is
maintained at a higher than normal level, the rates of deterioration will be
accelerated, resulting in a shorter than normal time for product failure. Because
the effects of extrinsic factors on deterioration can be quantified, the magnitude
of the acceleration can be calculated and the true shelf life of the product
under normal conditions calculated.
The need for ASLT of food products is simple: since many foods have
shelf lives of one year, evaluating the effect on shelf life of a change in theproduct, the process, or the packaging would require shelf life trials lasting
at least as long as the required shelf life of the product. Companies cannot
afford to wait for such long periods before knowing whether or not the new
product/process/packaging will give an adequate shelf life, because other
decisions have lead times of months and/or years. The use of ASLT in the
food industry is not as widespread as it might be, due in part to the lack of
basic data on the effect of extrinsic factors on the rates of deteriorative
reactions, in part to ignorance of the methodology required, and in part to a
skepticism of the advantages to be gained from using ASLT procedures.
Quality loss for most foods follows either a zero-order or first-order reaction.
Figure 13.1 showed the logarithm of shelf life versus temperature and the
inverse of absolute temperature. If only a small range of temperature is consid-
ered, the former shelf life plot generally fits the data for food products.
For a given extent of deterioration and reaction order, the rate constant is
inversely proportional to the time to reach some degree of quality loss. Thus
by taking the ratio of the shelf life between any two temperatures 10C (18F)
apart, the Q10 of the reaction can be found. This can be expressed by Equation(13.31):
Q10 =kT+10
kT=ST
ST+10
(13.29)
2000 by CRC Press, Inc.
8/3/2019 Shelf Life of Package Foods
20/25
where sT = shelf life at temperature TC and sT+10 = shelf life at temperature
(T+ 10)C assuming a linear shelf life plot. The effect ofQ10 on shelf life is
shown in Table 13.2, which illustrates the importance of accurate estimates
of the Q10 value when making shelf life predictions. Typical Q10 values for
foods have been reported as 1.1 to 4 for canned products, 1.5 to 10 fordehydrated products, and 3 to 40 for frozen products (Labuza, 1982).
A further use for Q10 values is illustrated inFigure 13.4, which depicts a
shelf life plot for a product that has at least 18 months shelf life at 23C
(73F). To determine the probable shelf life of the product at 40C (104F),
lines are drawn from the point corresponding to 18 months at 23C (73F) to
intersect a vertical line drawn at 40C (104F); the slope of each of the straight
lines so drawn is dictated by the Q10 value. Thus if the Q10 of the product
were 5, its shelf life at 40C (104F) would be 1 month, increasing to 5.4
months if the Q10 were 2. Such a plot is helpful in deciding how long an
ASLT is likely to run.
ASLT PROCEDURES
The following procedure should be adopted in developing a shelf life test
for a food product:
T Determine the microbiological safety and quality parameters for the
product.
T Select the key deteriorative reaction(s) that will cause quality loss and
thus consumer unacceptability in the product, and decide what tests
(sensory and/or instrumental) should be performed on the product
during the trial.
T Select the package to be used; often a range of packaging materials
will be tested so that the most cost-effective material can be selected.
T Select the extrinsic factors which are to be accelerated. Typical
TABLE 13.2. Effect of Q10 on Shelf Life.
Temperature Shelf Life (Weeks)
C F Q10 = 2 Q10 = 2.5 Q10 = 3 Q10 = 5
50 122 2* 2* 2* 2*
40 104 4 5 6 1030 86 8 12.5 18 50
20 68 16 31.3 54 4.8 years
* Arbitrarily set at 2 weeks at 50C (122F). Shelf lives at lower temperatures are calculated on thisarbitrary assumption.
Adapted from Robertson (1993), p. 368, by courtesy of Marcel Dekker, Inc.
2000 by CRC Press, Inc.
8/3/2019 Shelf Life of Package Foods
21/25
FIGURE 13.4 Hypothetical shelf life plot for various Q10s passing through a shelf life of 18
months at 23C (73F). Accelerated shelf life times (ASLT) are those required at 40C (104F)
for various Q10s. Reprinted from Robertson (1993), p. 369, by courtesy of Marcel Dekker, Inc.
storage conditions used for ASLT procedures are shown inTable
13.3, and it is usually necessary to select at least two.T Using a plot similar to that shown in Figure 13.4, determine how long
the product must be held at each test temperature. If no Q10 values are
known, then an open-ended ASLT will have to be conducted.
T Determine the frequency of the tests. A good rule of thumb (Labuza,
1985) is that the time interval between tests at any temperature below
the highest temperature should be no longer than
f2 = f1Q10/10
(13.30)
where f1 = the time between tests (e.g., days, weeks) at the highest test
temperature T1, f2 = the time between tests at any lower temperature T2, and
= the difference in degrees Celsius between T1 and T2.
2000 by CRC Press, Inc.
8/3/2019 Shelf Life of Package Foods
22/25
TABLE 13.3. Recommended Storage Conditions for ASLT.
Dry and Intermediate
Frozen Foods Moisture Foods Canned Foods
40C (40F) (control) 0C (32F) (control) 5C (41F) (control)
15C (10F) 23C (73F) (room temp.) 23C (73F) (room temp.)10C (14F) 30C (86F) 30C (86F)
5C (25F) 35C (95F) 35C (95F)
40C (104F) 40C (104F)
45C (113F)
Reprinted from Robertson (1993), p. 370, by courtesy of Marcel Dekker, Inc.
Thus if a product is held at 40C (104F) and tested once a month, then at
30C (86F) with a Q10 of 3, the product should be tested at least every
f2 = 1 3(10/10)
= 3 months
More frequent testing is desirable, especially if the Q10 is not accurately
known, and because at least six data points are needed to minimize statistical
errors, otherwise the confidence in s is significantly diminished.T Calculate the number of samples that must be stored at each test
condition, including those samples which will be held as controls.
T Begin the ASLTs, plotting the data as it comes to hand so that, if
necessary, the frequency of sampling can be increased or decreased as
appropriate.
T From each test storage condition, estimate kor s and construct
appropriate shelf life plots from which to estimate the potential shelf
life of the product under normal storage conditions. Provided that theshelf life plots indicate that the product shelf life is at least as long as
that desired by the company, then the product has a chance of
performing satisfactorily in distribution.
ASLT PROCEDURES FOR OXYGEN-SENSITIVE PRODUCTS
In all classical ASLT methods, temperature is the dominant acceleration
factor used, and its effect on the rate of lipid oxidation is best analyzed interms of the overall activation energy EA for lipid oxidation in fatty foods.
An inherent assumption in these tests is that Ea is the same in both the
presence and absence of antioxidants, although indications are that it is in
fact considerably lower in the latter case.
Other acceleration parameters which are used for shelf life are oxygen
2000 by CRC Press, Inc.
8/3/2019 Shelf Life of Package Foods
23/25
pressure, reactant contact, and the addition of catalysts. The effect of these
factors is generally much less important than that of temperature.
PROBLEMS IN THE USE OF ASLT PROCEDURES
The potential problems and theoretical errors which can arise in the use of
ASLT procedures include the following (Labuza and Schmidl, 1985):
T Error in analytical or sensory evaluation. Generally any analytical
measure should be done with a variability of less than 10% to
minimize prediction errors.
T As temperature rises, phase changes may occur (e.g., solid fat
becomes liquid) which can accelerate reactions, with the result that at
the lower temperature the actual shelf life will be shorter than
predicted.
T Carbohydrates in the amorphous state may crystallize at higher
temperatures, with the result that the predicted shelf life is shorter
than the actual shelf life at ambient conditions.
T Freezing control samples can result in reactants being concentrated
in the unfrozen liquid, creating a higher rate at certain temperatures
that is accounted for in the measured kvalue.
T If two reactions with different Q10 values cause quality loss in a food,the reaction with the higher Q10 may dominate at higher temperatures,
whereas at normal storage temperatures the reaction with the lower
Q10 may dominate, thus confounding the prediction.
T The aw of dry foods can increase with temperature, causing an
increase in the reaction rate for products of low aw in sealed packages.
This results in over-prediction of true shelf life at the lower
temperature.
T The solubility of gases (especially oxygen in fat or water) decreasesby almost 25% for each 10C (18F) rise in temperature. Thus an
oxidative reaction such as loss of ascorbic or linoleic acid can
decrease in rate if oxygen availability is the limiting factor. Therefore
at the higher temperature, the rate will be less than theoretical which
in turn will result in an underprediction of true shelf life at the normal
storage temperature.
T If the product is not placed in a totally impermeable pouch, storage in
high temperature low humidity cabinets generally enhances moistureloss, and this should increase the rate of quality loss compared to no
moisture change. This will result in a shorter predicted shelf life at the
lower temperature.
T If high enough temperatures are used, the product may actually be
cooked.
2000 by CRC Press, Inc.
8/3/2019 Shelf Life of Package Foods
24/25
Therefore, the use of ASLT to predict actual shelf life can be limited except
in the case of very simple chemical reactions. Consequently, food packaging
technologists should always confirm the ASLT results for a particular food
product by conducting shelf life tests under actual environmental conditions.
Once a relationship between ASLT and actual shelf life has been establishedfor a particular product, then ASLT can be used for that product when process
or package variables are to be evaluated.
BIBLIOGRAPHY
Anonymous. 1974. Shelf Life of Foods, Report by the Institute of Food Technologists Expert
Panel on Food Safety and Nutrition and the Committee on Public Information. Chicago, IL:
Institute of Food Technologists, August 1974. J. Food Sci., 39:861.Dethmers, A. E. 1979. Food Technology, 33(9):40.
Gacula, M. C. 1975. J. Food Sci., 40:399.
Gacula, M. C. and J. J. Kubala. 1975. J. Food Sci., 40:404.
Harte, B. R. and J. I. Gray. 1987. The Influence of Packaging on Product Quality, in Food
Product-Package Compatibility Proceedings. J. I. Gray, B.R. Harte, and J. Miltz, eds.Lancaster,
PA: Technomic Publishing Co., Inc., p. 17.
Heiss, R., U. Schrader, and G. R. Reinelt. 1980. The Influence of Diffusion and Solubility on
the Reaction of Oxygen in Compact Food, in Food Process Engineering, Vol. I, P. Linka,
Y. Malkki, J. Olkku, and J. Larinkari, eds. London, England: Applied Science Publishers Ltd.,Chapter 45.
Karel, M. 1974. Food Technology, 28(8):50.
Labuza, T. P. 1979. J. Food Sci., 44:1162.
Labuza, T. P. 1982. Shelf-Life Dating of Foods. Westport, CT: Food and Nutrition Press Inc.
Labuza, T. P. 1984. Moisture Sorption: Practical Aspects of Isotherm Measurement and Use.
St. Paul, MN: American Association of Cereal Chemists.
Labuza, T. P. 1985. An Integrated Approach to Food Chemistry: Illustrative Cases, in Food
Chemistry. O. R. Fennema, ed. New York: Marcel Dekker Inc., Chapter 16.
Labuza, T. P. and J. F. Kamman. 1983. Reaction Kinetics and Accelerated Tests Simulationas a Function of Temperature, in Computer-Aided Techniques in Food Technology. I. Saguy,
ed. New York: Marcel Dekker Inc., Chapter 4.
Labuza, T. P. and M. K. Schmidl. 1985. Food Technology, 39(9):57.
Labuza, T. P. and M. K. Schmidl. 1988. Cereal Foods World, 33(2):193.
Porter, W. L. 1981. Storage Life Prediction Under Noncontrolled Environmental Temperatures:
Product-Sensitive Environmental Call-Out, in Shelf-Life: A Key to Sharpening Your Competi-
tive Edge Proceedings. Washington, DC: Food Processors Institute, p. 1.
Quast, D. G. and M. Karel. 1972. J. Food Sci., 39:679.
Robertson, G. L. 1993. Food Packaging: Principles and Practice. New York: Marcel Dekker Inc.
Rosenfeld, P. E. 1984. Shelf Life Testing: Utilizing the Arrhenius Model to Characterize a
Distribution System, in Engineering and Food: Processing Applications, Vol. 2, B. M.
McKenna, ed. Essex, England: Elsevier Applied Science Publishers, Ltd., Chapter 16.
Samaniego-Esguerra, C. M. L., I. F. Boag, and G. L. Robertson. 1991. Lebensm.-Wiss. u.-
Technol., 24:53.
2000 by CRC Press, Inc.
8/3/2019 Shelf Life of Package Foods
25/25
Van Arsdel, W. B. 1969. Estimating Quality Change from a Known Temperature History, in
Quality and Stability of Frozen Foods. W. B. Van Arsdel, M. J. Copley, and R. L. Olson, eds.
New York: Wiley-Interscience, Chapter 10.
Van Den Berg, C. and S. Bruin. 1981. Water Activity and Its Estimation in Food Systems:
Theoretical Aspects, in Water Activity: Influences on Food Quality. L. B. Rockland and G.
F. Stewart, eds. New York: Academic Press, Chapter 1.